This project has been designed to elucidate and evaluate the metabolism of intracellular mycobacteria with a particular emphasis on those bacilli that are metabolically less active that persist in an immunocompetent host for long periods of time without progressing to active tuberculosis. Our initial approaches have focussed hard on oxygen regulated gene expression, particularly on the alpha crystallin homolog unique to slow-growing pathogens. We have also constructed a knockout in the cytochrome D system, a high-affinity, low-efficiency cytochrome utilized primarily under conditions of oxygen starvation. Construction of such knock-outs requires 6-9 months. This mutant shows some interesting phenotypes but these studies are at early stages. More recently we have begun to characterize knockouts in the relA system, that produces hyperphosphorylated guanine nucleotides as intracellular signaling molecules to marshall the so-called stringent response. This response may be involved in the bacterial adaptation to intramacrophage growth or perhaps more interestingly, in the transition to growth and persistence within the granuloma. Again studies are at an early stage but we have shown biochemically that mycobacteria utilize these ?alarmone? molecules to signal certain types of stress they encounter in the environment. A second component of this project involves the mycobacterial iron acquisition system. This project began during annotation of the genome of M. tuberculosis when we realized that the required enzymatic machinery for production of the 2-hydroxy- phenyloxazoline containing siderophores typical of slow-growing pathogenic mycobacteria appeared to be clustered in a single genetic operon. Gene disruption of this operon severely affected the ability of M. tuberculosis to grow under conditions of iron limitation as well as to grow within the environment of cultured human macrophages. M. tuberculosis produces two such siderophores related to one another but differing significantly in hydrophobicity so that one series is lipid soluble while the other series is water-soluble. This difference has given rise to the hypothesis that these two classes perform discrete functions, one operating as a traditional siderophore and competing for extracellular iron with molecules such as transferrin and the other operating as an ionophore, allowing transit through the lipophilic cell wall. We are now in a position to test this directly as this gene cluster encodes candidates for the acyl transferase which bifurcates the pathway and allows production of both classes. Our initial mutant disrupts both classes, establishing the biogenetic linkage between them and supporting our proposal for the bioynthetic pathway. The disrupted gene possesses a homolog in Pseudomonas aeruginosa, this gene pchS also forms a heterocyclic ring but instead of utilizing serine as the mycobacterial enzyme does resulting in formation of a phenyloxazoline ring system, it uses cysteine resulting in formation of a phenylthiazoline ring system. We have cloned the Pseudomonas enzyme into the Mycobacterial deletion strain and seen complementation for low-iron growth. There has been much speculation regarding the biological reasons for selecting one over the other, both atoms directly coordinate iron and affect the properties of the complex. We are also exploring this question synthetically and in collaboration with Dr. James DeVoss in Australia have synthesized the precursor molecule whose production is blocked in the mycobacterial system with either oxygen or sulfur present. We have shown that the oxy- or the thio- siderophore precursor can compensate under iron-limiting growth conditions. We are currently exploring in vivo conditions for differentiating the molecular function of the thia- and oxy- mycobactins.A final component of this project involves exploring biological effects of mycobacterial products on eukaryotic cells. Using recent clinical isolates and in collaboration with Dr. Gilla Kaplan at the Rockefeller University we have shown distinct biological effects of polar extracts of strains with distinct pathology and epidemiology. We are currently characterizing these differences but it is worth noting that another implication of the genome was that complex lipophilic molecules such as polyketides would be involved in such aspect of pathogenesis. Towards that end we are currently in the process of constructing defined genetic deletions of each of the two dozen polyketide synthase systems in a virulent strain of M. tuberculosis.